In the process of isolating and purifying an amino acid from an amino acid-containing aqueous solution, a purification method using an ion exchange resin is one of the most classical and representative unit operating methods. Since amino acids are amphoteric electrolytes, it is possible to use any of a cation exchange resin and an anion exchange resin for the purification thereof. In general, a strongly acidic cation exchange resin is most frequently employed. In this case, it is a general method that a fermentation liquid controlled to an appropriate hydrogen ion concentration is passed though a resin column and an adsorbed amino acid is eluted and recovered with an ammonium aqueous solution. A representative example is purification of L-lysine, which is one of amino acids for feedstuff, with a cation exchange resin. The ion exchange resin process has merits that the energy cost is smaller than a crystallization process or an electrodialysis process, and the facility cost is small in comparison with a membrane separation process.
The ion exchange resin process is roughly separated into an adsorption step and an elution step. The adsorption step is an adsorption step of adsorbing an amino acid (AA) on a resin (Resin) (the following formula (1)) and the elution step is a step of eluting the amino acid adsorbed on the resin using a predetermined eluent liquid (E) (the following formula (2)).Resin-E+AA→Resin-AA+E  (1)Resin-AA+E→Resin-E+AA  (2)
In such an ion exchange resin-treating operation, a liquid which is discharged from the adsorption step and contains substances not adsorbed on the ion exchange resin is called a breakthrough liquid, and a liquid which is discharged from the elution step and contains the amino acid is called an eluate. The problems of the ion exchange resin process are as follows. First, since the eluate contains not only the amino acid which is an objective substance but also the eluent, it is necessary to separate them. Since the ion exchange reaction is an equilibrium reaction, an excessive eluent relative to the amount of the amino acid in terms of mol ratio is necessarily required to elute the amino acid from the ion exchange resin. It can be represented by the following formula (3).Resin-AA+nE→Resin-E+AA+(n−1)E  (3)(wherein n>1)
Therefore, since the excess eluent ((n−1)E in the formula (3)) is mixed into the eluate, it is necessary to separate the amino acid from the eluent after the elution step. When the eluent is a volatile substance such as an ammonium component, it is easy to remove the excessive eluent by conducting mere concentration. However, in the case of a nonvolatile salt or the like, it is difficult to separate the salt and thereby the whole process becomes complex and the increase in a production cost is unavoidable.
The second problem relates to the separation of the eluent which is also present in the breakthrough liquid. Into the breakthrough liquid, not only non-adsorbed substances in the amino acid-containing aqueous solution, but also the eluent is eluted from the ion exchange resin by adsorbing the amino acid on the ion exchange resin, as shown by E in the right side of the formula (1). This eluent is usually converted into a by-product by forming a salt with a non-adsorbed substance in the fermentation liquid or the like. For example, when a lysine fermentation liquid is purified with a cation exchange resin using an ammonium component as an eluent, the sulfate ion in the fermentation liquid and the ammonium component in the eluent are combined to form ammonium sulfate.
There may be no problem when such a by-product is a volatile substance or the like and is easily recovered. However, since the by-product is usually a stable salt, it still remains until the end of the process. Therefore, in the process for purifying a fermentation liquid using the ion exchange process, generation of a commensurate by-product is unavoidable.
As one method for overcoming the defect of such an ion exchange resin process, there has been known a method of ion-exchanging an amino acid adsorbed with an anion exchange resin using as an eluent liquid, an aqueous carbonic acid solution obtained by dissolving carbon dioxide gas in water under predetermined pressure (Pasztor, Zs.; Chanel, S.; Grevillot, G. Elution of amino acids from anion exchange columns by means of an aqueous solution of carbon dioxide under pressure. AIChE Annual Meeting, Miami, Nov. 15-19, 1998). In this method, when the obtained eluate is released to normal pressure, the eluent present in the eluate, i.e., a carbonate ion is converted into gas to vaporize, and hence the eluent does not remain in the eluate as shown in formula (4). Therefore, it becomes unnecessary to separate the eluent from the objective substance after the elution step, and thus the purification of the eluate is simplified.Resin-AA−+nHCO3−→Resin-HCO3−+AA−+(n−1)OH−+(n−1)CO2↑(wherein n>1)  (4)
There are three problems in this method. First, in the adsorption step of re-adsorbing an amino acid on a carbonate-type anion exchange resin after the elution of the amino acid, carbon dioxide gas foams as shown in formulae (5) and (6).Resin-HCO3−+AA−→Resin-AA−+nHCO3−  (5)HCO3−→OH−+CO2↑  (6)
When the carbon dioxide gas is generated in the column, smooth ion exchange is not achieved since channeling of the liquid occurs. Second, in order to re-adsorb the amino acid on the anion exchange resin, it is necessary to regenerate the carbonate-type anion exchange resin to the OH-type anion exchange resin once, and hence the process becomes complex. Industrially, it is preferable to adsorb the amino acid directly on the carbonate-type anion exchange resin. Third, in the elution step, amino acids capable of being eluted are limited to neutral amino acids. The elution reaction is conducted at around the neutral hydrogen ion concentration. For example, when the amino acid is an acidic amino acid, since the acidic amino acid is in the form of an anion, the amino acid once eluted from the resin is immediately re-adsorbed on the resin. Usual ion exchange reaction proceeds microscopically with repeating the elution and re-adsorption. However, when an aqueous carbonic acid solution wherein carbon dioxide gas is dissolved in water is used as an eluent liquid, the carbonate ion concentration in the eluent liquid is low and the carbonate ion necessary for re-elution of the re-adsorbed amino acid is not necessarily sufficient. As a result, a sufficient elution cannot be achieved in the case of an acidic amino acid.
On the other hand, with regard to the recovery of eluents such as ammonium radical and carbonate radical, JP-A-55-139815 discloses a method of obtaining carbon dioxide in a remaining liquid phase as gaseous carbon dioxide after fractional distillation of an ammonium component contained in an aqueous solution.